# eCommerce AI Agent Roadmap — Phasing

Companion to **How to Build an AI Agent Roadmap for an eCommerce Business**.

Post 40 ([`ai-automation-opportunity-score.md`](ai-automation-opportunity-score.md)) ranks **which** workflow. This artifact sequences **when**. Demo rows are a worksheet, not a FactualMinds engagement.

```text
Phase 1 Quick wins
  → Phase 2 AI intelligence
  → Phase 3 Operational agents
  → Phase 4 Cross-system automation
  → Phase 5 Advanced agent operations
```

## Prioritization factors (each 1–5)

| Factor | 5 looks like | Veto |
| ------ | ------------ | ---- |
| **Business impact** | Money, CX load, or trust | Impact ≠ permission to write |
| **Data readiness** | Joinable IDs + freshness SLA | Missing IDs → stay in Phase 1 reads |
| **Integration complexity** | Named OpenAPI already exists | Email-only 3PL → Draft only |
| **Risk** | Read-only or idempotent notify | Payment, ATP, live price, account |
| **Automation potential** | High volume, messy language | Flow already closes it |

## Phase map

| Phase | What ships | Typical workflows | Exit gate |
| ----- | ---------- | ----------------- | --------- |
| **1 Quick wins** | Read tools, one harness, goldens | WISMO, policy lookup, returns **recommend** | Evidence_tool on every answer |
| **2 Intelligence** | Investigation, not writes | Daily brief, store Q&A, data quality flags | Schema-constrained output |
| **3 Operational agents** | One governed write domain | Exception recommend, catalog **draft**, HITL refund | Cedar LOG_ONLY → ENFORCE |
| **4 Cross-system** | Same session, multiple systems | Order↔WMS↔helpdesk; ERP PO **draft** | Identity per audience |
| **5 Advanced** | Supervisor only if hops needed | Control tower, specialist team | Hop caps; not Swarm-by-default |

Skip a phase only if the exit gate of the previous phase is already true **in production**, not in a slide.

## Sample 12-month sequence (replace)

| Quarter | Phase | Staff | Do not |
| ------- | ----- | ----- | ------ |
| Q1 | 1 | WISMO + returns recommend | Inventory qty write |
| Q2 | 2 | Daily brief + quality monitors | Multi-agent supervisor |
| Q3 | 3 | Exception agent + HITL refund | Live price Execute |
| Q4 | 4 or stop | ERP/CRM tools if APIs exist | Level 5 because a vendor demoed it |

## Related

- Opportunity score: [`ai-automation-opportunity-score.md`](ai-automation-opportunity-score.md)
- Maturity: [`ai-agent-maturity-model.md`](ai-agent-maturity-model.md)
- Monday: [`monday-checklist.md`](monday-checklist.md)